Alt-BEAM Archive

Message #03816



To: beam@sgiblab.sgi.com
From: Dennison dennlill@buffnet.net
Date: Thu, 27 May 1999 20:58:59 -0400
Subject: [alt-beam] Re: BEAManT 6PBS special- Poppers



I too have been working on these more advanced photovores. My work stems
from my earlyer work With BitBots, (bicore photovores) (Qiuck note, if you
use only one motor with the latest suneater you have an effecient solar
powered head) Anyway, I came up with the XOR motor driver circuit also,
(INdependently of Mark T I might add) It was part of a project to
incorporate increased manuverability into the robot using Nu's. A nifty
nifty idea, I'm kinda glad I thought of it! :-) (and several other people at
the same time, althoug mark probably was first!) Anyway some of the work
I've been tinkering with lately deals with looking for that super effecient
look. Brian had a point about not deep discharging caps. The idea is to let
the bot fill up the cap and then just sip of the top. That way, the cap
voltage is always high,and the lost power is easy (and quick) to replace. As
a result, one circuit I was trying was a Sem-Nested Bicore. In this, A PM1
SE controlls the total on control, IE: The first time the bot runs it need
to sit in the sun for a few minutes to completly fill his caps up. Then the
PM1 turns on a Bicore, one of these bicore outputs controls the power to
another bicore (effectivly turning the second bicore on and off, not just
de-enabling the inverters. By not powering the chip I belive I save the
current that would be slowely consumed by the 240 in 'sleep' mode.) The
amount of time that the second bicore is opperational (the second bicore
controlls the photovore motions itself) is then determined by the first
bicore. This bicore acts as a timer, keeping the second bicore active for
only certian amounts of time. The time durration isn't constant, a CDS cell
alters the time of the first 'control' bicore so that the robot runs for
(slightly) longer periods in bright light, and charges quicker, and in low
light it runs for signifigantly less time and charges for a greater period
of time. However the bot always appear quick and nimble because the PM1 SE
keeps the overall cap voltage high, if the bicore controll ever lets the
second bicore draw the power below acceptable levels, the PM1 shuts off the
entire bicore nest to allow the whole shebang to 'recharge'.

Complicated?

Dennison

>I love it! I've been playing around for 3 weeks off and on trying to do
this
>very thing. No wonder progress was slow, I was dabbling in Mark's arena
(like
>I could ever come close, right). I still haven't got it to work. I never
>thought of limiting the phases like that. I've been thinking, sleep a long
>time, move for a fair amount. Somewhat boring at times. This puts a whole
new
>spin on it for me. I love this list!
>Jim


------------------------------------------------------------------------

eGroups.com home: http://www.egroups.com/group/alt-beam
http://www.egroups.com - Simplifying group communications



Home